Continual Test-Time Adaptation

Follow this topic Ask about this topic

Momentum

8 papers in the last four weeks, up 167% on the four weeks before. 0.1% of all new papers.

Jul 13Week of Sep 28

Latest papers 45

All topics

Self-Generated Feedback Destabilizes Test-Time Training: A Causal Decomposition of Long-Horizon Adaptation
Oct 4, 2026Cheng Luo, Bing Li, Bernard GhanemTest-Time Adaptation Continual Test-Time Adaptation
King Abdullah University of Science and Technology (KAUST)
When to Adapt: Multi-Signal Domain Shift Detection for Efficient Training-Free Adaptation in Open-Vocabulary Segmentation
Sep 29, 2026Michele Antonazzi, Alejandra C. Hernandez, José Araujo +2Distribution Shift Continual Test-Time Adaptation
EECS and Digital Futures, KTH Royal Institute of Technology, Stockholm, Sweden · Ericsson Research, Ericsson AB, Stockholm, Sweden
Not Every Correction Helps: Gain-Guided Continual Test-Time Adaptation
Sep 29, 2026Youjia Zhang, Huiling Liu, Soyun Choi +2Adaptive Inference Test-Time Adaptation
Sungkyunkwan University · Nanyang Technological University
SafeCoEvo: Co-Evolving Safety Harnesses and Guards for LLM Agents at Test-Time
Sep 29, 2026Yu Cheng, Yongkang Hu, Shuaijie Ma +12Continual Test-Time Adaptation LLM Agents
East China Normal University · Shanghai Innovation Institute · Xiamen University +4
Transferable Mass Spectrum Prediction via Reference-Guided Test-time Specialization
Sep 28, 2026Yunhua Zhong, Runting Li, Yifan Li +5Test-Time Adaptation Continual Test-Time Adaptation
The Hong Kong University of Science and Technology (Guangzhou) · The University of Hong Kong · South China University of Technology +3
One Rollout Is All You Get: Fully Test-Time Adaptation for GUI Agents
Sep 28, 2026Ziqiang Wang, Li Gu, Zhixiang Chi +6Test-Time Adaptation Continual Test-Time Adaptation
Concordia University · Mila – Qu´ebec AI Institute · University of Toronto +1
Continuous Online Fault Detection for Mobile Robots via Adaptive Edge Models
Sep 24, 2026Jordan Levy, Nicolas Verstaevel, Vincent Talon +1Robot Failure Detection Online Anomaly Detection
IRIT, Université Toulouse Capitole, Toulouse, France · TwinswHeel, Soben, France
SPeaR: Test-Time Adaptation with Steering Primitives for Realigning Representations
Sep 21, 2026Muhammad Sudipto Siam Dip, Ali EtemadTest-Time Adaptation Continual Test-Time Adaptation
Department of Electrical and Computer Engineering Queen’s University, Canada
Rethinking the Teacher-Student Framework for Test-Time Adaptation
Sep 2, 2026Damian Sójka, Marc Masana, Bartłomiej Twardowski +1Test-Time Adaptation Teacher-Student Learning
Poznan University of Technology · Akces NCBR · Graz University of Technology +4
When Does Online Adaptation Pay on the Edge? A Leakage-Free Evaluation of Warmup, Learning-Rate Selection, and Resource Trade-offs for Time-Series Forecasting
Sep 1, 2026Takumi Fujimoto, Hiroaki NishiMultivariate Time Series Forecasting Learning Rate Scheduling
School of Science for Open and Environmental Systems, Graduate School of Science and Technology, Keio University, Yokohama, Japan
Continual Test-Time Adaptation via Entropy Sensitivity-Guidance in Strict Online Setting
Aug 30, 2026Chandler Timm C. Doloriel, Yunbei Zhang, Muhammad Salman Siddiqui +4Distribution Shift Robustness Test-Time Adaptation
Faculty of Science and Technology (REALTEK), Norwegian University of Life Sciences (NMBU) · Tulane University
A2TTA: Anchored-and-Agile Test-Time Adaptation for Evolving Traffic Sensor Networks
Jul 28, 2026Du Yin, Xiachong Lin, Yue Tan +4Intelligent Transportation Systems Continual Test-Time Adaptation
University of New South Wales · Sydney, NSW, Australia · Griffith University +7
Source-Free Controlled Adaptation of Teachers for Continual Test-Time Adaptation
Jul 26, 2026Anurag Roy, Riddhiman Moulick, Vinay Kumar Verma +2Source-Free Domain Adaptation Domain Adaptation
IIT Kharagpur · IBM Research · Amazon +1
Black-Mamba: Biologically-Inspired Leaky Accumulation for Conceptual Knowledge under Distribution Drift
Jul 21, 2026Giuseppe Soriano, Nicola Tonellotto, Alberto GottaTest-Time Adaptation Concept Drift
University of Pisa National Research Council (CNR) Pisa, Italy · University of Pisa Pisa, Italy · National Research Council (CNR) Pisa, Italy
On-Device Adaptive Battery Power Prediction for Electric Vehicles
Jul 10, 2026Avik Bhatnagar, Anton Paule, Tobias Schuermann +2On-Device Training Continual Test-Time Adaptation
FZI Research Center for Information Technology, University of Tübingen Germany
Continual Test-Time Adaptation in Computer Vision: Methods, Benchmarks, and Future Directions
Jul 9, 2026Sarthak Kumar Maharana, Shambhavi Mishra, Yunbei Zhang +6Source-Free Domain Adaptation Distribution Shift Robustness
The University of Texas at Dallas, USA · LIVIA ETS Montreal, ILLS International Laboratory on Learning Systems (ILLS), McGill - ÉTS - MILA - CNRS - Université Paris-Saclay - CentraleSupélec · Tulane University, USA +2
Open-Ended Scenario Reasoning for Specialist Model Adaptation
Jul 7, 2026Youcheng Zong, Runda Jia, Ranmeng Lin +2Continual Test-Time Adaptation
College of Information Science and Engineering, Northeastern University, Shenyang 110004, China
Continual Model Merging with Test-Time Adaptation for Whole-Slide Image Analysis
Jul 6, 2026Duc-Thanh Le, Doanh C. Bui, Maï K. Nguyen +1Continual Test-Time Adaptation Histopathology Image Classification
University of Information Technology · 2Viet Nam National University Ho Chi Minh City · 3ETIS (UMR 8051), CY Cergy Paris University, ENSEA, CNRS, France
NeuroOnline: Bridging Pretraining and Online Adaptation for EEG Foundation Models
Jul 4, 2026Weibin Li, Wendu Li, Yushan You +2Electroencephalography Continual Test-Time Adaptation
1Southern University of Science and Technology
AdaJEPA: An Adaptive Latent World Model
Jun 30, 2026Ying Wang, Oumayma Bounou, Yann LeCun +1Continual Test-Time Adaptation World Model-Based Planning
New York University · AMI Labs
BP-TTA: Balanced and Prototype-Guided Test-Time Adaptation in Dynamic Scenarios
Jun 30, 2026Shaoyang Huang, Yashi Zhu, Yichen Yu +3Class-Imbalanced Learning Test-Time Adaptation
College of Computer Science, Sichuan University, Chengdu, 610065, China · School of Artificial Intelligence, Sichuan University, Chengdu, 610065, China
DriftGuard: Safety-Aware Multi-Monitor Detection and Selective Adaptation for Evolving Toxicity Moderation
Jun 27, 2026Yuting Xin, Hanyu Cai, Binqi Shen +2Concept Drift Continual Test-Time Adaptation
Department of Information and Decision Sciences University of Minnesota Minneapolis, USA · Department of Industrial Engineering and Management Sciences Northwestern University Evanston, USA · Fuqua School of Business Duke University Durham, USA +1
Forget, Anticipate and Adapt: Test Time Training for Long Videos
Jun 25, 2026Rajat Modi, Sebastian Noel, Xin Liang +1Continual Test-Time Adaptation Test-Time Training
Institute of Artificial Intelligence, University of Central Florida
Hitting a Moving Target: Test-Time Adaptation for AI Text Detection under Continual Distribution Shift
Jun 23, 2026Kevin Ren, Manish Raghavan, Nikhil GargAI-Generated Text Detection Distribution Shift
Cornell Tech · MIT
Adapt Only When It Pays: Budgeted Decision-Loss Priority for Delayed Online Time-Series Adaptation
Jun 23, 2026Xibai WangTest-Time Adaptation Continual Test-Time Adaptation
NeuroQuant Labs Limited Hong Kong SAR, China
PromptDyG: Test-Time Prompt Adaptation on Dynamic Graphs
Jun 22, 2026Guoguo Ai, Chaoxi Niu, Hui Yan +3Continual Test-Time Adaptation Dynamic Graph Learning
School of Computer Science and Engineering, Nanjing University of Science and Technology, China · School of Computing and Information Systems, Singapore Management University, Singapore · Faculty of Data Science, City University of Macau, Macau, China +3
Distill Once, Adapt Life-Long: Exploring Dataset Distillation for Continual Test-Time Adaptation
Jun 18, 2026Hyun-Kurl Jang, Jihun Kim, Hyeokjun Kweon +1Source-Free Domain Adaptation Continual Test-Time Adaptation
KAIST, Visual Intelligence Lab · Chung-Ang University, FoVLab
Validation-Gated Multi-Agent Governance for Online Adaptation of Thermal-Hydraulic Surrogate Models under Operating-Regime Shift
Jun 2, 2026Doyeong Lim, Seungyoon Lee, In Cheol BangNeural Surrogate Modeling Continual Test-Time Adaptation
Department of Nuclear Engineering, Ulsan National Institute of Science and Technology (UNIST), Ulsan, 44919, Republic of Korea
TimeBlocks: Foundational and Continual Time-Series Blockbase -- Extended Version
Jun 1, 2026David Campos, Bin Yang, Tung Kieu +3Continual Test-Time Adaptation Multi-Task Learning
Aalborg University, Denmark · University of Amsterdam Netherlands · East China Normal University China +1
Private and Stable Test-Time Adaptation with Differential Privacy
Jun 1, 2026Zefeng Li, Qiaoyue Tang, Mathias Lecuyer +1Continual Test-Time Adaptation Privacy-Preserving ML
University of British Columbia · Vector Institute

Oct 4, 2026, cs.CLJ/K move · Enter open · S save

Self-Generated Feedback Destabilizes Test-Time Training: A Causal Decomposition of Long-Horizon Adaptation

Cheng Luo, Bing Li, Bernard Ghanem

King Abdullah University of Science and Technology (KAUST)

Test-Time Adaptation Continual Test-Time Adaptation Self-Training for Language Models Test-Time Training

Open paper AskPDF arXiv

Test-time training (TTT) lets a model store information in its weights during inference. When the model learns from its own output, however, each update also changes the model that generates the next training example. Across 128K-token streams, retaining generated-text updates worsens prediction on independent human-written text with three TTT-E2E model configurations (labeled 125M, 760M, and 3B). The same failure occurs when Adam updates Qwen3-4B's existing weights. The same update mechanisms can improve on real text, so writing itself is not the failure. Three matched comparisons trace the causal pathway. Fixed Generation removes over 98% of the damage at 125M and 760M by using a frozen model to generate training chunks. Recorded Replay separates the loss caused by reading degraded text from the additional loss stored by updating on it. A paired one-update comparison then shows the local conflict: an update predicts its source better but new real text worse. This cost grows after Closed Loop adaptation, with a few trajectories accounting for most large failures. Finally, Settlement evaluates the candidate state on independent real text before commitment. It leaves mean endpoint gaps of 0.07 and -0.02 nats at 125M and 760M while retaining real-text adaptation. These results motivate checking prediction on independent evidence before retaining an update.

Momentum

Related topics

Latest papers 45